strange peak near end of elution

[s]

Hello,

I am running GPC with a Waters Ultrahydrogel column and can not seem to get rid of a double peak that shows up near the end of my runs. The peak shows up on my refractive index detector but not my fluorescence detector. My eluent is phosphate buffer. I make all my samples in the same eluent buffer. The Vo for my column is ~6min. and a Vt of ~15min. The flow rate is 0.8ml/min and the eluent sparged with helium almost continuously.

On RI detector, here are some findings:
(1) Peak is a positive double peak when running with a blank (i.e., elution buffer)

(2) Peak is a negative double peak when running a 1 to 10 dilution of a FITC labelled sample

(3) Peak is a positive double peak when running non-diluted version of FITC labelled sample

(4) During equilibration mode, there is no peak that shows up at all.

**I do get a valid peak on fluorescence detector but don't know how to interpret this ghost peak on RI detector.

I have been flushing my manual injector with water in between sample runs using a needle port cleaner so I am pretty sure it is not contamination between samples. Could the refractive index detector cell be dirty? If it was, why don't I see a ghost peak while in equilibration mode? Also, I don't see how this can be a solvent peak b/c it shows up near Vt and also, it is quite a significant peak.

Any ideas for what this is or how to get rid of it would be helpful.

S

[Noser222]

The reason you see a peak in your RI and not your fluorescence is that is where small molecules (excess FITC), salts, and water will elute. If you injected a sample of pure water you'd see a huge negative peak. I wouldn't worry about what goes on in that part of the chromatogram as long as you are getting valid results for your analyte of interest.

[s]

Noser222,

Thank you for the response. I worry about getting a response b/c the double peak is so close to my Vt. Any ideas on how to get rid of it? Sometimes it is higher than my analylate of interest too, which is annoying.

S

[HW Mueller]

If you called what you are doing SEC you may have realized that your molecules are too small for this column to be excluded enough to come sufficiently ahead of the "normal" small molecules. There might be one possibility to increase your separation, namely, if your analytes are macromolecules which can be manipulated as to effective size with mobile phase changes (you would have to increase the effective size to make them come earlier).

[Uwe Neue]

What are the MW of your analytes of interest, and which pore size are you using? If what appears to be a salt peak is too close to your analyte peak, you may need to reduce the pore size of the packing, or add a column with a smaller pore size.

[s]

The molecular weight of my fluorescently labelled sample (FITC-Ficoll) is 400KD. It gives a very broad curve from 6min to 13.4min on fluoro detector and then the mysterious peak comes out at around 13.6min only on the RI detector. Note, the sample concentration is too low to show up on RI detector but unfortunately, the mystery peak shows up on RI.

Here is an image of RI detection with different samples (including water and PBS):

[s]

The pore size is 10um. Waters Ultrahydrogel 500.

S

[HW Mueller]

...hmmm... don´t know about Waters´ SEC, but that 10µm is probably for the particle size?

[Alfred88]

Dear s [sargum]:
From your picture, I believe that your Vt is ~18 min, not ~15min. (Can any experts of SEC comment on this?) So, your peak of interest is still in “safeâ€

[Andyp]

I often see additional "unknown" peaks within this region of a SEC separation. I've monitored these and have found them to change significantly during the course of a multi-sample run. In the case of SEC using organic solvents. They also change significantly when the injection solvent has been left open to the atmosphere, and when the solvent has been deliberately exposed to high humidity. It would appear, in my case, that the unknown peaks are due to differences in the RI between running solvent & injection solvent, and possibly due to dissolved atmospheric moisture, gasses etc???

I have been successful in reducing the size of these peaks but have never been successful in eliminating them. To reduce them, you need to be using mobile phase taken directly from your reservoir for sample prep.

Hope this is useful

[HW Mueller]

One need not to worry about non-excluded peaks (salts, small molecules...), except, perhaps if they are absent. One needs to choose a column which definately does exclusion of the analytes. We still don´t know if the right column (pore size, recommendet mol weight for different macromolecule shapes) was used here. (A 10µm pore should be for elephants, not molecules?)

[Andyp]

I strongly suspect that 10um is the particle size.